Allometry in the terminal velocity - dispersal architecture relationship explains variation in dispersal and offspring provisioning strategies in wind dispersed Asteraceae species

Competition can simultaneously favour high dispersal ability (to transport offspring to more favourable habitats) and large seed size (to maximise offspring provisioning). In wind dispersed Asteraceae species, seeds are enclosed within an achene with hair-like projections from the achene form a pappus that acts as a parachute to aid in dispersal. There is potentially an allometric relationship between terminal velocity and pappus to achene volume ratio (dispersal architecture), with changes in dispersal architecture resulting in disproportionately high or low impacts on terminal velocity. We tested the hypothesis that competition induces shifts in dispersal architecture depending on the allometric relationship between terminal velocity and dispersal architecture. We estimated dispersal architecture of diaspores from seven wind dispersed Asteraceae species from environments with low and high neighbour densities. We also estimated diaspore terminal velocity for a subset of these species by recording drop time in a 2 m tube. Diaspores of one species had dispersal architecture promoting higher dispersal under high neighbour density, diaspores of two species had dispersal architecture promoting lower dispersal under high neighbour density, and dispersal architecture was not significantly different between high and low density environments for four of the species. Species showed a common allometric relationship between terminal velocity and dispersal architecture. The allometric relationship predicts dispersal architecture changes across environments differing in neighbour density. Species with dispersal architecture promoting higher dispersal under high neighbour density do so where small increases in dispersal architecture yield large decreases in terminal velocity. Our research suggests that the nature of allometric relationships between traits can help to explain allocation strategies across environments.